lukasplu (Luka Pluzynski)

upvoted a paper 12 days ago

DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videos

Paper • 2409.02095 • Published 13 days ago • 31

upvoted 2 papers about 1 month ago

Task-oriented Sequential Grounding in 3D Scenes

Paper • 2408.04034 • Published Aug 7 • 8

Transformer Explainer: Interactive Learning of Text-Generative Models

Paper • 2408.04619 • Published Aug 8 • 152

upvoted a paper about 2 months ago

BetterDepth: Plug-and-Play Diffusion Refiner for Zero-Shot Monocular Depth Estimation

Paper • 2407.17952 • Published Jul 25 • 27

upvoted a paper 3 months ago

Florence-2: Advancing a Unified Representation for a Variety of Vision Tasks

Paper • 2311.06242 • Published Nov 10, 2023 • 77

upvoted a collection 3 months ago

Florence

Collection

9 items • Updated Jul 11 • 153

upvoted a paper 3 months ago

Depth Anything V2

Paper • 2406.09414 • Published Jun 13 • 91

upvoted an article 4 months ago

Article

A Dive into Pretraining Strategies for Vision-Language Models

Feb 3, 2023

• 32

upvoted 2 articles 5 months ago

Article

seemore: Implement a Vision Language Model from Scratch

By

•

Jun 23

• 56

Article

Welcome Llama 3 - Meta's new open LLM

Apr 18

• 271

upvoted a collection 5 months ago

Llama 3

Collection

8 items • Updated Apr 18 • 12

upvoted an article 5 months ago

Article

Vision Language Models Explained

Apr 11

• 172

upvoted 2 papers 6 months ago

DepthFM: Fast Monocular Depth Estimation with Flow Matching

Paper • 2403.13788 • Published Mar 20 • 16

SceneScript: Reconstructing Scenes With An Autoregressive Structured Language Model

Paper • 2403.13064 • Published Mar 19 • 31

upvoted 8 papers 8 months ago

CARFF: Conditional Auto-encoded Radiance Field for 3D Scene Forecasting

Paper • 2401.18075 • Published Jan 31 • 8

Rephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling

Paper • 2401.16380 • Published Jan 29 • 46

Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities

Paper • 2401.14405 • Published Jan 25 • 11

Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild

Paper • 2401.13627 • Published Jan 24 • 71

upvoted 4 papers 9 months ago

Unsupervised Universal Image Segmentation

Paper • 2312.17243 • Published Dec 28, 2023 • 19

ControlRoom3D: Room Generation using Semantic Proxy Rooms

Paper • 2312.05208 • Published Dec 8, 2023 • 8

Zero-Shot Metric Depth with a Field-of-View Conditioned Diffusion Model

Paper • 2312.13252 • Published Dec 20, 2023 • 27

Mosaic-SDF for 3D Generative Models

Paper • 2312.09222 • Published Dec 14, 2023 • 15

upvoted a paper 10 months ago

ControlMat: A Controlled Generative Approach to Material Capture

Paper • 2309.01700 • Published Sep 4, 2023 • 13

Luka Pluzynski

AI & ML interests

Organizations

lukasplu's activity

DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videos

Task-oriented Sequential Grounding in 3D Scenes

Transformer Explainer: Interactive Learning of Text-Generative Models

BetterDepth: Plug-and-Play Diffusion Refiner for Zero-Shot Monocular Depth Estimation

Florence-2: Advancing a Unified Representation for a Variety of Vision Tasks

Florence

Depth Anything V2

A Dive into Pretraining Strategies for Vision-Language Models

seemore: Implement a Vision Language Model from Scratch

Welcome Llama 3 - Meta's new open LLM

Llama 3

Vision Language Models Explained

DepthFM: Fast Monocular Depth Estimation with Flow Matching

SceneScript: Reconstructing Scenes With An Autoregressive Structured Language Model

CARFF: Conditional Auto-encoded Radiance Field for 3D Scene Forecasting

Rephrasing the Web: A Recipe for Compute and Data-Efficient Language Modeling

Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities

Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild

Single-View 3D Human Digitalization with Large Reconstruction Models

EmerDiff: Emerging Pixel-level Semantic Knowledge in Diffusion Models

Make-A-Shape: a Ten-Million-scale 3D Shape Model

SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities

Unsupervised Universal Image Segmentation

ControlRoom3D: Room Generation using Semantic Proxy Rooms

Zero-Shot Metric Depth with a Field-of-View Conditioned Diffusion Model

Mosaic-SDF for 3D Generative Models

ControlMat: A Controlled Generative Approach to Material Capture